SUMMARY/ABSTRACT Epigenetic remodeling is increasingly recognized as a driver of carcinogenesis and a promising drug target for cancer therapy. Histone deacetylase inhibitors (HDIs) are among the most prominent epigenetic drugs being tested in over 500 clinical trials against a variety of cancers with two compounds already approved. Despite their universal anticancer efficacy, their mechanism of action is not clear. It is currently assumed that, by inhibiting histone deacetylase (HDAC), HDIs alter expression of genes involved in cell growth or survival. However, gene expression profiling of HDI-treated cells revealed minimal changes and distinct patterns among different HDIs. Many HDACs, including HDAC1, 2, 3, 6, and their associated transcriptional coregulators, have been shown to function as tumor suppressors. There is clearly a gap of knowledge on how HDIs work. Our recent findings challenge the current view of HDIs. We found that liver-specific knockout of HDAC3 upregulates lipogenic target genes and causes hepatic steatosis, both of which can be rescued by enzyme-dead mutants of HDAC3. We further found that HDIs do not upregulate target genes despite causing histone hyperacetylation. Such dissociation between enzyme activity and HDAC function questions whether HDAC is de facto target of HDI, especially considering that HDAC3 is responsible for the enzyme activity of class IIa HDACs (HDAC4, 5, 7, and 9) that do not possess intrinsic catalytic activity. The view of HDACs as oncogene is also at odds with their tumor suppressor role in retinoblastoma protein and estrogen/androgen receptor signaling pathways. HDIs are designed to chelate the zinc ion in the catalytic pocket of HDACs3, and therefore could target hundreds of other zinc-dependent metalloproteins. We hypothesize that non-HDAC proteins are de facto targets of HDI for its anticancer efficacy. We will tests the hypothesis by determining whether HDAC enzymatic activity is required for cellular response to HDIs; characterizing the role of HDI-interacting non-HDAC proteins in anticancer effects of HDIs; and determining the role HDAC enzymatic activity in liver carcinogenesis and hepatic response to HDIs in vivo. Our hypothesis, if proven correct, is a conceptual advance that will revolutionize our strategy in development of these promising anticancer drugs. Our study also directly tests the ?histone code? hypothesis on an unprecedented scale in mammals, and will have broad impact in the field of epigenetics and gene transcriptional regulation.